aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl
diff options
context:
space:
mode:
authorSergey M․ <dstftw@gmail.com>2018-10-29 23:29:05 +0700
committerSergey M․ <dstftw@gmail.com>2018-10-29 23:32:45 +0700
commit2943397e8701d3dcd28433e485e50459fdbda62a (patch)
treec092f2f3440470d78413bba0873ee36e131a2c51 /youtube_dl
parent9c4a83a1bec9d7abd066a89be40e62dd36ffa67d (diff)
[aparat] Improve extraction and extract more metadata (closes #17445, closes #18008)
Diffstat (limited to 'youtube_dl')
-rw-r--r--youtube_dl/extractor/aparat.py89
1 files changed, 57 insertions, 32 deletions
diff --git a/youtube_dl/extractor/aparat.py b/youtube_dl/extractor/aparat.py
index 780439e17..883dcee7a 100644
--- a/youtube_dl/extractor/aparat.py
+++ b/youtube_dl/extractor/aparat.py
@@ -4,6 +4,7 @@ from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
int_or_none,
+ merge_dicts,
mimetype2ext,
url_or_none,
)
@@ -12,59 +13,83 @@ from ..utils import (
class AparatIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
- _TEST = {
+ _TESTS = [{
'url': 'http://www.aparat.com/v/wP8On',
'md5': '131aca2e14fe7c4dcb3c4877ba300c89',
'info_dict': {
'id': 'wP8On',
'ext': 'mp4',
'title': 'تیم گلکسی 11 - زومیت',
- 'age_limit': 0,
+ 'description': 'md5:096bdabcdcc4569f2b8a5e903a3b3028',
+ 'duration': 231,
+ 'timestamp': 1387394859,
+ 'upload_date': '20131218',
+ 'view_count': int,
},
- # 'skip': 'Extremely unreliable',
- }
+ }, {
+ # multiple formats
+ 'url': 'https://www.aparat.com/v/8dflw/',
+ 'only_matching': True,
+ }]
def _real_extract(self, url):
video_id = self._match_id(url)
- # Note: There is an easier-to-parse configuration at
- # http://www.aparat.com/video/video/config/videohash/%video_id
- # but the URL in there does not work
- webpage = self._download_webpage(
- 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
- video_id)
+ # Provides more metadata
+ webpage = self._download_webpage(url, video_id, fatal=False)
+
+ if not webpage:
+ # Note: There is an easier-to-parse configuration at
+ # http://www.aparat.com/video/video/config/videohash/%video_id
+ # but the URL in there does not work
+ webpage = self._download_webpage(
+ 'http://www.aparat.com/video/video/embed/vt/frame/showvideo/yes/videohash/' + video_id,
+ video_id)
- file_list = self._parse_json(
+ options = self._parse_json(
self._search_regex(
- r'var options\s*=\s*JSON\.parse\(\'([^\']+)\'\)', webpage,
- 'file list'),
+ r'options\s*=\s*JSON\.parse\(\s*(["\'])(?P<value>(?:(?!\1).)+)\1\s*\)',
+ webpage, 'options', group='value'),
video_id)
- title = file_list['plugins']['sabaPlayerPlugin']['title']
+ player = options['plugins']['sabaPlayerPlugin']
formats = []
- for list in file_list['plugins']['sabaPlayerPlugin']['multiSRC']:
- for item in list:
+ for sources in player['multiSRC']:
+ for item in sources:
+ if not isinstance(item, dict):
+ continue
file_url = url_or_none(item.get('src'))
if not file_url:
continue
- ext = mimetype2ext(item.get('type'))
- label = item.get('label')
- formats.append({
- 'url': file_url,
- 'ext': ext,
- 'format_id': label or ext,
- 'height': int_or_none(self._search_regex(
- r'(\d+)[pP]', label or '', 'height', default=None)),
- })
- self._sort_formats(formats)
+ item_type = item.get('type')
+ if item_type == 'application/vnd.apple.mpegurl':
+ formats.extend(self._extract_m3u8_formats(
+ file_url, video_id, 'mp4',
+ entry_protocol='m3u8_native', m3u8_id='hls',
+ fatal=False))
+ else:
+ ext = mimetype2ext(item.get('type'))
+ label = item.get('label')
+ formats.append({
+ 'url': file_url,
+ 'ext': ext,
+ 'format_id': 'http-%s' % (label or ext),
+ 'height': int_or_none(self._search_regex(
+ r'(\d+)[pP]', label or '', 'height',
+ default=None)),
+ })
+ self._sort_formats(
+ formats, field_preference=('height', 'width', 'tbr', 'format_id'))
+
+ info = self._search_json_ld(webpage, video_id, default={})
- thumbnail = file_list['poster']
+ if not info.get('title'):
+ info['title'] = player['title']
- return {
+ return merge_dicts(info, {
'id': video_id,
- 'title': title,
- 'thumbnail': thumbnail,
- 'age_limit': self._family_friendly_search(webpage),
+ 'thumbnail': url_or_none(options.get('poster')),
+ 'duration': int_or_none(player.get('duration')),
'formats': formats,
- }
+ })